Not if self driving vehicles are also adopted by the railways.

:-)

But I think the main people use railways is because they don't want to be driving, or aren't able to park at the other end. In exchange for that, they put up with the cost of the tickets and the inconvenience of not travelling door-to-door.

But self-driving cars solve almost all of the issues that trains are intended to fix.

Though, as Joe Newman pointed out in response to this on Facebook, they probably won't be as fast as the long-distance high-speed trains. Don't think I'd invest in any railway stock at the moment, though :-)

Railways are a lot more efficient than electric cars in terms of energy consumption per kilo shipped. I think they will still be used heavily for long distance freight. They are probably more efficient in terms of passenger throughput as well, although probably less so than right now since self-driving cars can drive very close together. So we might still use subways in congested urban areas. That said, the convenience of door-to-door transport may be a more important factor than the overall efficiency of the system.

Well, I have a car and love to travel door-to-door. But I have to commute daily by rail as I can't afford to live where I work, it's more reliable than driving, and the station is located much more conveniently in my work destination than the park and ride. I do put up with shabby trains and the occasional very bad delay, but my actual cost of living is cheaper and the elements of chaos that can disrupt a car journey (individual breakdowns, crashes, congestion to think of three) seem to be in shorter supply for me. Moreover I was looking at it more that whatever self-driving cars are meant to be good at, the railways will want those things too. Are they likely to stand by and watch that happen without investing in their own version of it and rolling it out to make use of the infrastructure that railways already have (rolling stock, maintenance teams, track, terminii, journey monitoring, buildings and in many areas miles and miles of electrification)?

If you are referring to breakdowns, crashes or congestion, then I can't see how the autonomous car, i.e. something approximating a personal vehicle will be immune to any of these issues. Every piece of technology ever, probably from the flint onwards, can always be made less useful by being a servant of wetware; "nothing is idiot-proof". Whatever the tech, even with skilled users the operator can break it (or find it suddenly broken), exhaust the supply of it (or whatever is needed to make it work), attempt to do something with it which means it won't work, attempt to improve it (introducing obsolescence), or expect it to be used in the same way by everyone. It's the last idea that dents much of the optimism I feel I ought to have for autonomous vehicles. Somewhere in the vision of this-being-better is, I think, an assumption that travel means the same thing to everyone, and it doesn't. We can reduce it to the idea of "getting from a to b" but it doesn't factor in the idea that I might want to leave it as late as possible for me get from a to b, whilst you might want to get from a to b much faster than me. Bob might want to get from a to b whilst being drunk/asleep/with his mistress, whilst Ted wants to get from a to b but can't be bothered to plan this in advance and faffs about when it is time and holds both me and you up. Meanwhile Hugo wants to go from b to a because he left what he needed to bring to b back at a and Conrad thought he wanted to go from a to b but should really have gone to c and will only realise this just when it becomes a massive inconvenience to everyone else. Of course autonomous vehicles could in theory be made to travel closer together and deal with speed and safety in different ways to human-controlled vehicles, but I think the intent of travel is more significant for "chaos" than the application of travel. A breakdown of any vehicle, autonomous or not, will mess up the journey of others around it. Meanwhile we even get congestion in networks delivering packets of data, so I still think it'll be a reality for the driverless vehicle. Meanwhile, all of the characters above will be in autonomous vehicles of differing ages, makes and specification, with all the variance that we can expect from such a scenario (upgrades, performance, Beta-vs-VHS, premium and economy, brand loyalty and so on). Even if we expect constant, mass re-use of identical autonomous vehicles with no personal ownership, I can't help but see something like the situation we have with supermarket trolleys being a reality for the shared, autonomous vehicle. So in spite of being "all the same", some will always have wonky wheels, some will have remnants of whatever whoever used it before you did with it, some will be abandoned far away from all the others, some will be stolen or maliciously abused, some will be overloaded when used, and some days you'll only have the choice of one that is too big/too small for what you want to do and everything will rattle around annoyingly. Now, autonomous vehicles may well make significant improvements in transporting non-human cargo and introduce production-line efficiency to those things-that-we-do-that-aren't-travel. But these things might run much better away from the chaos of human life, somewhere standard, unified and with pre-agreed routes where high speeds can be attained. You know, some sort of system with tracks and signals and depots :P

Thanks Jim - a long and thoughtful response!

I wrote a long reply, but decided it would be better as a separate post. Coming soon...

Q